To predict rainfall, our proposed model architecture combines the Convolutional Neural Network (CNN), which uses the ResNet-152 pre-training model, with the Recurrent Neural Network (RNN), which uses the Long Short-term Memory Network (LSTM) layer, for model training. By encoding the cloud images through CNN, we extract the image feature vectors in the training process and train the vectors and meteorological data as the input of RNN. After training, the accuracy of the prediction model can reach up to 82%. The result has proven not only the outperformance of our proposed rainfall prediction method in terms of cost and prediction time, but also its accuracy and feasibility compared with general prediction methods.
{"title":"Machine Learning-based Short-term Rainfall Prediction from Sky Data","authors":"Fu Jie Tey, Tin-Yu Wu, Jiann-Liang Chen","doi":"10.1145/3502731","DOIUrl":"https://doi.org/10.1145/3502731","url":null,"abstract":"To predict rainfall, our proposed model architecture combines the Convolutional Neural Network (CNN), which uses the ResNet-152 pre-training model, with the Recurrent Neural Network (RNN), which uses the Long Short-term Memory Network (LSTM) layer, for model training. By encoding the cloud images through CNN, we extract the image feature vectors in the training process and train the vectors and meteorological data as the input of RNN. After training, the accuracy of the prediction model can reach up to 82%. The result has proven not only the outperformance of our proposed rainfall prediction method in terms of cost and prediction time, but also its accuracy and feasibility compared with general prediction methods.","PeriodicalId":435653,"journal":{"name":"ACM Transactions on Knowledge Discovery from Data (TKDD)","volume":"239 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115601148","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Recently, learning and mining from data streams with incremental feature spaces have attracted extensive attention, where data may dynamically expand over time in both volume and feature dimensions. Existing approaches usually assume that the incoming instances can always receive true labels. However, in many real-world applications, e.g., environment monitoring, acquiring the true labels is costly due to the need of human effort in annotating the data. To tackle this problem, we propose a novel incremental Feature spaces Learning with Label Scarcity (FLLS) algorithm, together with its two variants. When data streams arrive with augmented features, we first leverage the margin-based online active learning to select valuable instances to be labeled and thus build superior predictive models with minimal supervision. After receiving the labels, we combine the online passive-aggressive update rule and margin-maximum principle to jointly update the dynamic classifier in the shared and augmented feature space. Finally, we use the projected truncation technique to build a sparse but efficient model. We theoretically analyze the error bounds of FLLS and its two variants. Also, we conduct experiments on synthetic data and real-world applications to further validate the effectiveness of our proposed algorithms.
{"title":"Incremental Feature Spaces Learning with Label Scarcity","authors":"Shilin Gu, Yuhua Qian, Chenping Hou","doi":"10.1145/3516368","DOIUrl":"https://doi.org/10.1145/3516368","url":null,"abstract":"Recently, learning and mining from data streams with incremental feature spaces have attracted extensive attention, where data may dynamically expand over time in both volume and feature dimensions. Existing approaches usually assume that the incoming instances can always receive true labels. However, in many real-world applications, e.g., environment monitoring, acquiring the true labels is costly due to the need of human effort in annotating the data. To tackle this problem, we propose a novel incremental Feature spaces Learning with Label Scarcity (FLLS) algorithm, together with its two variants. When data streams arrive with augmented features, we first leverage the margin-based online active learning to select valuable instances to be labeled and thus build superior predictive models with minimal supervision. After receiving the labels, we combine the online passive-aggressive update rule and margin-maximum principle to jointly update the dynamic classifier in the shared and augmented feature space. Finally, we use the projected truncation technique to build a sparse but efficient model. We theoretically analyze the error bounds of FLLS and its two variants. Also, we conduct experiments on synthetic data and real-world applications to further validate the effectiveness of our proposed algorithms.","PeriodicalId":435653,"journal":{"name":"ACM Transactions on Knowledge Discovery from Data (TKDD)","volume":"14 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-06-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124814616","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
One key objective of artificial intelligence involves the continuous adaptation of machine learning models to new tasks. This branch of continual learning is also referred to as lifelong learning (LL), where a major challenge is to minimize catastrophic forgetting, or forgetting previously learned tasks. While previous work on catastrophic forgetting has been focused on vision problems; this work targets time-series data. In addition to choosing an architecture appropriate for time-series sequences, our work addresses limitations in previous work, including the handling of distribution shifts in class labels. We present multi-objective learning with three loss functions to minimize catastrophic forgetting, prediction error, and errors in generalizing across label shifts, simultaneously. We build a multi-task autoencoder network with a hierarchical convolutional recurrent architecture. The proposed method is capable of learning multiple time-series tasks simultaneously. For cases where the model needs to learn multiple new tasks, we propose sequential learning, starting with tasks that have the best individual performances. This solution was evaluated on four benchmark human activity recognition datasets collected from mobile sensing devices. A wide set of baseline comparisons is performed, and an ablation analysis is run to evaluate the impact of the different losses in the proposed multi-objective method. The results demonstrate an up to 4% performance improvement in catastrophic forgetting compared to the use of loss functions in state-of-the-art solutions while demonstrating minimal losses compared to upper bound methods of traditional fine-tuning (FT) and multi-task learning (MTL).
{"title":"Multi-objective Learning to Overcome Catastrophic Forgetting in Time-series Applications","authors":"Reem A. Mahmoud, Hazem M. Hajj","doi":"10.1145/3502728","DOIUrl":"https://doi.org/10.1145/3502728","url":null,"abstract":"One key objective of artificial intelligence involves the continuous adaptation of machine learning models to new tasks. This branch of continual learning is also referred to as lifelong learning (LL), where a major challenge is to minimize catastrophic forgetting, or forgetting previously learned tasks. While previous work on catastrophic forgetting has been focused on vision problems; this work targets time-series data. In addition to choosing an architecture appropriate for time-series sequences, our work addresses limitations in previous work, including the handling of distribution shifts in class labels. We present multi-objective learning with three loss functions to minimize catastrophic forgetting, prediction error, and errors in generalizing across label shifts, simultaneously. We build a multi-task autoencoder network with a hierarchical convolutional recurrent architecture. The proposed method is capable of learning multiple time-series tasks simultaneously. For cases where the model needs to learn multiple new tasks, we propose sequential learning, starting with tasks that have the best individual performances. This solution was evaluated on four benchmark human activity recognition datasets collected from mobile sensing devices. A wide set of baseline comparisons is performed, and an ablation analysis is run to evaluate the impact of the different losses in the proposed multi-objective method. The results demonstrate an up to 4% performance improvement in catastrophic forgetting compared to the use of loss functions in state-of-the-art solutions while demonstrating minimal losses compared to upper bound methods of traditional fine-tuning (FT) and multi-task learning (MTL).","PeriodicalId":435653,"journal":{"name":"ACM Transactions on Knowledge Discovery from Data (TKDD)","volume":"30 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-06-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124918689","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Monitoring systems have hundreds or thousands of distributed sensors gathering and transmitting real-time streaming data. The early detection of events in these systems, such as an earthquake in a seismic monitoring system, is the base for essential tasks as warning generations. To detect such events is usual to compute pairwise correlation across the disparate signals generated by the sensors. Since the data sources (e.g., sensors) are spatially separated, it is essential to consider the lagged correlation between the signals. Besides, many applications require to process a specific band of frequencies depending on the event’s type, demanding a pre-processing step of filtering before computing correlations. Due to the high speed of data generation and a large number of sensors in these systems, the operations of filtering and lagged cross-correlation need to be efficient to provide real-time responses without data losses. This article proposes a technique named FilCorr that efficiently computes both operations in one single step. We achieve an order of magnitude speedup by maintaining frequency transforms over sliding windows. Our method is exact, devoid of sensitive parameters, and easily parallelizable. Besides our algorithm, we also provide a publicly available real-time system named Seisviz that employs FilCorr in its core mechanism for monitoring a seismometer network. We demonstrate that our technique is suitable for several monitoring applications as seismic signal monitoring, motion monitoring, and neural activity monitoring.
{"title":"Combining Filtering and Cross-Correlation Efficiently for Streaming Time Series","authors":"Sheng Zhong, Vinicius M. A. Souza, A. Mueen","doi":"10.1145/3502738","DOIUrl":"https://doi.org/10.1145/3502738","url":null,"abstract":"Monitoring systems have hundreds or thousands of distributed sensors gathering and transmitting real-time streaming data. The early detection of events in these systems, such as an earthquake in a seismic monitoring system, is the base for essential tasks as warning generations. To detect such events is usual to compute pairwise correlation across the disparate signals generated by the sensors. Since the data sources (e.g., sensors) are spatially separated, it is essential to consider the lagged correlation between the signals. Besides, many applications require to process a specific band of frequencies depending on the event’s type, demanding a pre-processing step of filtering before computing correlations. Due to the high speed of data generation and a large number of sensors in these systems, the operations of filtering and lagged cross-correlation need to be efficient to provide real-time responses without data losses. This article proposes a technique named FilCorr that efficiently computes both operations in one single step. We achieve an order of magnitude speedup by maintaining frequency transforms over sliding windows. Our method is exact, devoid of sensitive parameters, and easily parallelizable. Besides our algorithm, we also provide a publicly available real-time system named Seisviz that employs FilCorr in its core mechanism for monitoring a seismometer network. We demonstrate that our technique is suitable for several monitoring applications as seismic signal monitoring, motion monitoring, and neural activity monitoring.","PeriodicalId":435653,"journal":{"name":"ACM Transactions on Knowledge Discovery from Data (TKDD)","volume":"75 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-05-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130132557","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Time-varying dynamic Bayesian network (TVDBN) is essential for describing time-evolving directed conditional dependence structures in complex multivariate systems. In this article, we construct a TVDBN model, together with a score-based method for its structure learning. The model adopts a vector autoregressive (VAR) model to describe inter-slice and intra-slice relations between variables. By allowing VAR parameters to change segment-wisely over time, the time-varying dynamics of the network structure can be described. Furthermore, considering some external information can provide additional similarity information of variables. Graph Laplacian is further imposed to regularize similar nodes to have similar network structures. The regularized maximum a posterior estimation in the Bayesian inference framework is used as a score function for TVDBN structure evaluation, and the alternating direction method of multipliers (ADMM) with L-BFGS-B algorithm is used for optimal structure learning. Thorough simulation studies and a real case study are carried out to verify our proposed method’s efficacy and efficiency.
{"title":"Segment-Wise Time-Varying Dynamic Bayesian Network with Graph Regularization","authors":"Xingxuan Yang, Chen Zhang, Baihua Zheng","doi":"10.1145/3522589","DOIUrl":"https://doi.org/10.1145/3522589","url":null,"abstract":"Time-varying dynamic Bayesian network (TVDBN) is essential for describing time-evolving directed conditional dependence structures in complex multivariate systems. In this article, we construct a TVDBN model, together with a score-based method for its structure learning. The model adopts a vector autoregressive (VAR) model to describe inter-slice and intra-slice relations between variables. By allowing VAR parameters to change segment-wisely over time, the time-varying dynamics of the network structure can be described. Furthermore, considering some external information can provide additional similarity information of variables. Graph Laplacian is further imposed to regularize similar nodes to have similar network structures. The regularized maximum a posterior estimation in the Bayesian inference framework is used as a score function for TVDBN structure evaluation, and the alternating direction method of multipliers (ADMM) with L-BFGS-B algorithm is used for optimal structure learning. Thorough simulation studies and a real case study are carried out to verify our proposed method’s efficacy and efficiency.","PeriodicalId":435653,"journal":{"name":"ACM Transactions on Knowledge Discovery from Data (TKDD)","volume":"77 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-05-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"113958294","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pengfei Ma, Youxi Wu, Y. Li, Lei Guo, He Jiang, Xingquan Zhu, X. Wu
As a novel deep learning model, gcForest has been widely used in various applications. However, current multi-grained scanning of gcForest produces many redundant feature vectors, and this increases the time cost of the model. To screen out redundant feature vectors, we introduce a hashing screening mechanism for multi-grained scanning and propose a model called HW-Forest which adopts two strategies: hashing screening and window screening. HW-Forest employs perceptual hashing algorithm to calculate the similarity between feature vectors in hashing screening strategy, which is used to remove the redundant feature vectors produced by multi-grained scanning and can significantly decrease the time cost and memory consumption. Furthermore, we adopt a self-adaptive instance screening strategy called window screening to improve the performance of our approach, which can achieve higher accuracy without hyperparameter tuning on different datasets. Our experimental results show that HW-Forest has higher accuracy than other models, and the time cost is also reduced.
{"title":"HW-Forest: Deep Forest with Hashing Screening and Window Screening","authors":"Pengfei Ma, Youxi Wu, Y. Li, Lei Guo, He Jiang, Xingquan Zhu, X. Wu","doi":"10.1145/3532193","DOIUrl":"https://doi.org/10.1145/3532193","url":null,"abstract":"As a novel deep learning model, gcForest has been widely used in various applications. However, current multi-grained scanning of gcForest produces many redundant feature vectors, and this increases the time cost of the model. To screen out redundant feature vectors, we introduce a hashing screening mechanism for multi-grained scanning and propose a model called HW-Forest which adopts two strategies: hashing screening and window screening. HW-Forest employs perceptual hashing algorithm to calculate the similarity between feature vectors in hashing screening strategy, which is used to remove the redundant feature vectors produced by multi-grained scanning and can significantly decrease the time cost and memory consumption. Furthermore, we adopt a self-adaptive instance screening strategy called window screening to improve the performance of our approach, which can achieve higher accuracy without hyperparameter tuning on different datasets. Our experimental results show that HW-Forest has higher accuracy than other models, and the time cost is also reduced.","PeriodicalId":435653,"journal":{"name":"ACM Transactions on Knowledge Discovery from Data (TKDD)","volume":"75 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-05-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121643267","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
A. Bechini, Alessandro Bondielli, José Luis Corcuera Bárcena, P. Ducange, F. Marcelloni, Alessandro Renda
In the last years, there has been an ever-increasing interest in profiling various aspects of city life, especially in the context of smart cities. This interest has become even more relevant recently when we have realized how dramatic events, such as the Covid-19 pandemic, can deeply affect the city life, producing drastic changes. Identifying and analyzing such changes, both at the city level and within single neighborhoods, may be a fundamental tool to better manage the current situation and provide sound strategies for future planning. Furthermore, such fine-grained and up-to-date characterization can represent a valuable asset for other tools and services, e.g., web mapping applications or real estate agency platforms. In this article, we propose a framework featuring a novel methodology to model and track changes in areas of the city by extracting information from online newspaper articles. The problem of uncovering clusters of news at specific times is tackled by means of the joint use of state-of-the-art language models to represent the articles, and of a density-based streaming clustering algorithm, properly shaped to deal with high-dimensional text embeddings. Furthermore, we propose a method to automatically label the obtained clusters in a semantically meaningful way, and we introduce a set of metrics aimed at tracking the temporal evolution of clusters. A case study focusing on the city of Rome during the Covid-19 pandemic is illustrated and discussed to evaluate the effectiveness of the proposed approach.
{"title":"A News-Based Framework for Uncovering and Tracking City Area Profiles: Assessment in Covid-19 Setting","authors":"A. Bechini, Alessandro Bondielli, José Luis Corcuera Bárcena, P. Ducange, F. Marcelloni, Alessandro Renda","doi":"10.1145/3532186","DOIUrl":"https://doi.org/10.1145/3532186","url":null,"abstract":"In the last years, there has been an ever-increasing interest in profiling various aspects of city life, especially in the context of smart cities. This interest has become even more relevant recently when we have realized how dramatic events, such as the Covid-19 pandemic, can deeply affect the city life, producing drastic changes. Identifying and analyzing such changes, both at the city level and within single neighborhoods, may be a fundamental tool to better manage the current situation and provide sound strategies for future planning. Furthermore, such fine-grained and up-to-date characterization can represent a valuable asset for other tools and services, e.g., web mapping applications or real estate agency platforms. In this article, we propose a framework featuring a novel methodology to model and track changes in areas of the city by extracting information from online newspaper articles. The problem of uncovering clusters of news at specific times is tackled by means of the joint use of state-of-the-art language models to represent the articles, and of a density-based streaming clustering algorithm, properly shaped to deal with high-dimensional text embeddings. Furthermore, we propose a method to automatically label the obtained clusters in a semantically meaningful way, and we introduce a set of metrics aimed at tracking the temporal evolution of clusters. A case study focusing on the city of Rome during the Covid-19 pandemic is illustrated and discussed to evaluate the effectiveness of the proposed approach.","PeriodicalId":435653,"journal":{"name":"ACM Transactions on Knowledge Discovery from Data (TKDD)","volume":"131 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-04-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132346378","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Leonardo Pellegrina, Cyrus Cousins, Fabio Vandin, Matteo Riondato
“I’m an MC still as honest” – Eminem, Rap God We present MCRapper, an algorithm for efficient computation of Monte-Carlo Empirical Rademacher Averages (MCERA) for families of functions exhibiting poset (e.g., lattice) structure, such as those that arise in many pattern mining tasks. The MCERA allows us to compute upper bounds to the maximum deviation of sample means from their expectations, thus it can be used to find both (1) statistically-significant functions (i.e., patterns) when the available data is seen as a sample from an unknown distribution, and (2) approximations of collections of high-expectation functions (e.g., frequent patterns) when the available data is a small sample from a large dataset. This flexibility offered by MCRapper is a big advantage over previously proposed solutions, which could only achieve one of the two. MCRapper uses upper bounds to the discrepancy of the functions to efficiently explore and prune the search space, a technique borrowed from pattern mining itself. To show the practical use of MCRapper, we employ it to develop an algorithm TFP-R for the task of True Frequent Pattern (TFP) mining, by appropriately computing approximations of the negative and positive borders of the collection of patterns of interest, which allow an effective pruning of the pattern space and the computation of strong bounds to the supremum deviation. TFP-R gives guarantees on the probability of including any false positives (precision) and exhibits higher statistical power (recall) than existing methods offering the same guarantees. We evaluate MCRapper and TFP-R and show that they outperform the state-of-the-art for their respective tasks.
{"title":"MCRapper: Monte-Carlo Rademacher Averages for Poset Families and Approximate Pattern Mining","authors":"Leonardo Pellegrina, Cyrus Cousins, Fabio Vandin, Matteo Riondato","doi":"10.1145/3532187","DOIUrl":"https://doi.org/10.1145/3532187","url":null,"abstract":"“I’m an MC still as honest” – Eminem, Rap God We present MCRapper, an algorithm for efficient computation of Monte-Carlo Empirical Rademacher Averages (MCERA) for families of functions exhibiting poset (e.g., lattice) structure, such as those that arise in many pattern mining tasks. The MCERA allows us to compute upper bounds to the maximum deviation of sample means from their expectations, thus it can be used to find both (1) statistically-significant functions (i.e., patterns) when the available data is seen as a sample from an unknown distribution, and (2) approximations of collections of high-expectation functions (e.g., frequent patterns) when the available data is a small sample from a large dataset. This flexibility offered by MCRapper is a big advantage over previously proposed solutions, which could only achieve one of the two. MCRapper uses upper bounds to the discrepancy of the functions to efficiently explore and prune the search space, a technique borrowed from pattern mining itself. To show the practical use of MCRapper, we employ it to develop an algorithm TFP-R for the task of True Frequent Pattern (TFP) mining, by appropriately computing approximations of the negative and positive borders of the collection of patterns of interest, which allow an effective pruning of the pattern space and the computation of strong bounds to the supremum deviation. TFP-R gives guarantees on the probability of including any false positives (precision) and exhibits higher statistical power (recall) than existing methods offering the same guarantees. We evaluate MCRapper and TFP-R and show that they outperform the state-of-the-art for their respective tasks.","PeriodicalId":435653,"journal":{"name":"ACM Transactions on Knowledge Discovery from Data (TKDD)","volume":"43 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-04-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116047692","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
People’s location data are continuously tracked from various devices and sensors, enabling an ongoing analysis of sensitive information that can violate people’s privacy and reveal confidential information. Synthetic data have been used to generate representative location sequences yet to maintain the users’ privacy. Nonetheless, the privacy-accuracy tradeoff between these two measures has not been addressed systematically. In this article, we analyze the use of different synthetic data generation models for long location sequences, including extended short-term memory networks (LSTMs), Markov Chains (MC), and variable-order Markov models (VMMs). We employ different performance measures, such as data similarity and privacy, and discuss the inherent tradeoff. Furthermore, we introduce other measurements to quantify each of these measures. Based on the anonymous data of 300 thousand cellular-phone users, our work offers a road map for developing policies for synthetic data generation processes. We propose a framework for building data generation models and evaluating their effectiveness regarding those accuracy and privacy measures.
{"title":"Synthesis of Longitudinal Human Location Sequences: Balancing Utility and Privacy","authors":"Maya Benarous, Eran Toch, I. Ben-Gal","doi":"10.1145/3529260","DOIUrl":"https://doi.org/10.1145/3529260","url":null,"abstract":"People’s location data are continuously tracked from various devices and sensors, enabling an ongoing analysis of sensitive information that can violate people’s privacy and reveal confidential information. Synthetic data have been used to generate representative location sequences yet to maintain the users’ privacy. Nonetheless, the privacy-accuracy tradeoff between these two measures has not been addressed systematically. In this article, we analyze the use of different synthetic data generation models for long location sequences, including extended short-term memory networks (LSTMs), Markov Chains (MC), and variable-order Markov models (VMMs). We employ different performance measures, such as data similarity and privacy, and discuss the inherent tradeoff. Furthermore, we introduce other measurements to quantify each of these measures. Based on the anonymous data of 300 thousand cellular-phone users, our work offers a road map for developing policies for synthetic data generation processes. We propose a framework for building data generation models and evaluating their effectiveness regarding those accuracy and privacy measures.","PeriodicalId":435653,"journal":{"name":"ACM Transactions on Knowledge Discovery from Data (TKDD)","volume":"16 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-04-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130036552","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Anomaly detection is an essential task for quality management in smart manufacturing. An accurate data-driven detection method usually needs enough data and labels. However, in practice, there commonly exist newly set-up processes in manufacturing, and they only have quite limited data available for analysis. Borrowing the name from the recommender system, we call this process a cold-start process. The sparsity of anomaly, the deviation of the profile, and noise aggravate the detection difficulty. Transfer learning could help to detect anomalies for cold-start processes by transferring the knowledge from more experienced processes to the new processes. However, the existing transfer learning and multi-task learning frameworks are established on task- or domain-level relatedness. We observe instead, within a domain, some components (background and anomaly) share more commonality, others (profile deviation and noise) not. To this end, we propose a more delicate component-level transfer learning scheme, i.e., decomposition-based hybrid transfer learning (DHTL): It first decomposes a domain (e.g., a data source containing profiles) into different components (smooth background, profile deviation, anomaly, and noise); then, each component’s transferability is analyzed by expert knowledge; Lastly, different transfer learning techniques could be tailored accordingly. We adopted the Bayesian probabilistic hierarchical model to formulate parameter transfer for the background, and “L2,1+L1”-norm to formulate low dimension feature-representation transfer for the anomaly. An efficient algorithm based on Block Coordinate Descend is proposed to learn the parameters. A case study based on glass coating pressure profiles demonstrates the improved accuracy and completeness of detected anomaly, and a simulation demonstrates the fidelity of the decomposition results.
{"title":"Profile Decomposition Based Hybrid Transfer Learning for Cold-Start Data Anomaly Detection","authors":"Ziyue Li, Haodong Yan, F. Tsung, Ke Zhang","doi":"10.1145/3530990","DOIUrl":"https://doi.org/10.1145/3530990","url":null,"abstract":"Anomaly detection is an essential task for quality management in smart manufacturing. An accurate data-driven detection method usually needs enough data and labels. However, in practice, there commonly exist newly set-up processes in manufacturing, and they only have quite limited data available for analysis. Borrowing the name from the recommender system, we call this process a cold-start process. The sparsity of anomaly, the deviation of the profile, and noise aggravate the detection difficulty. Transfer learning could help to detect anomalies for cold-start processes by transferring the knowledge from more experienced processes to the new processes. However, the existing transfer learning and multi-task learning frameworks are established on task- or domain-level relatedness. We observe instead, within a domain, some components (background and anomaly) share more commonality, others (profile deviation and noise) not. To this end, we propose a more delicate component-level transfer learning scheme, i.e., decomposition-based hybrid transfer learning (DHTL): It first decomposes a domain (e.g., a data source containing profiles) into different components (smooth background, profile deviation, anomaly, and noise); then, each component’s transferability is analyzed by expert knowledge; Lastly, different transfer learning techniques could be tailored accordingly. We adopted the Bayesian probabilistic hierarchical model to formulate parameter transfer for the background, and “L2,1+L1”-norm to formulate low dimension feature-representation transfer for the anomaly. An efficient algorithm based on Block Coordinate Descend is proposed to learn the parameters. A case study based on glass coating pressure profiles demonstrates the improved accuracy and completeness of detected anomaly, and a simulation demonstrates the fidelity of the decomposition results.","PeriodicalId":435653,"journal":{"name":"ACM Transactions on Knowledge Discovery from Data (TKDD)","volume":"36 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-04-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126670152","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：